应用前白蛋白检测营养风险筛查2002、主观全面评定和微小营养评定的灵敏度和特异度：探索性、前瞻性、多中心临床研究

目的 探索应用前白蛋白水平为参考标准检测营养风险筛查2002( NRS 2002)、主观全面评定(SGA)及微小营养评定(MNA)的灵敏度和特异度.方法 自2010年11月至2011年3月,在3个医院的肿瘤科住院患者中采用连续定点抽样方法,将符合入选条件的126例食管癌患者纳入本研究.于入院次日分别采用NRS 2002、SGA及MNA工具由经过培训的营养师、肿瘤内科医师进行营养筛查或评定.以前白蛋白低于180 g/L为营养不良(不足)的参考标准,分析上述3种工具的灵敏度、特异度.结果 126例符合入选条件的患者均完成NRS 2002、SGA及MNA的筛查或评定.NRS 2002、SGA及MNA检出的有营养风险和营养不良(不足)率分别为67.0％、46.2％、64.8％.以与临床结局有关的前白蛋白为参考标准,NRS 2002、SGA及MNA的灵敏度分别为78.2％、58.2％、89.1％,特异度分别为50.0％、72.2％、27.8％.结论 3种工具中,NRS 2002检出的是营养风险,而SGA和MNA检出的是营养不良(不足),3种工具之间有不同特点.由于单一的前白蛋白不是一个适当的参考标准,不能用于3种工具的比较.但NRS 2002有关临床结局,适用于住院患者.需要进一步进行以临床结局为检测标准的多中心研究.

The use of prealbumin as index to compare sensitivities and specificities of Nutritional Risk Screening 2002, Subjective Global Assessment, Mini Nutritional Assessment: a prospective multicenter clinical studies

Objective To explore sensitivities and specificities of Nutritional Risk Screening 2002 ( NRS 2002),Subjective Global Assessment (SGA),and Mini Nutritional Assessment (MNA) in the detection of blood prealbumin level.Methods Total 126 patients were consecutively enrolled from the oncology departments of three hospitals from November,2010 to March,2011.Nutritional screenings or assessments were performed on the next morning of hospital admission by nutritionists and medical oncologists using NRS 2002,SGA,and MNA.Prealbumin ( ＜ 180 g/L) was used as the reference standard of malnutrition (undernutrition) for assessing the sensitivities/specificities of these tools.Results All 126 patients completed nutritional screening or assessment using the above three tools.The prevalence of malnutrition (undernutrition) determined by NRS 2002,SGA and MNA was 67.0％,46.2％,and 64.8％,respectively.With prealbumin in relation to clinical outcome as the reference standard,the sensitivities ofNRS 2002,SGA,and MNA were 78.2％,58.2％ and 89.1％,and the specificities were 50.0％,72.2％,and 27.8％,respectively.Conclusions Among these three tolls,NRS 2002 detects the nutrition risks,whereas SGA and MNA detect malnutrition.Since prealbumin is not an appropriate reference standard,it can not be used for the comparison among the three tools.Nevertheless,NRS 2002 reflects the clinical outcomes,and therefore is suitable for inpatients.